Differences in educational attainment

What sort of people do well at school? This report from DfES sheds some fascinating light. Some key points:
1. There are big class differences in attainment. Only 4.6% of children in the most deprived quartile are high attainers (ie, in the top 10% of performers) at Key Stage 2 (age 11). But 16.7% of children from the least deprived quartile are high attainers. The figures are similar for 16-year-olds.
2. Ethnic differences are even bigger than class differences. At age 11, 25.4% of Chinese pupils are high attainers. That compares to 3.9% of Black Caribbeans, 10.5% of British whites, 13.4% of Indians and 4.4% of Pakistanis. Again, the differences are similar at 16.
3. High attainment doesn't always persist. Only 45% of kids who were in the top 10% at age 11 were in the top 10% at 16.
Is this an argument against grammar schools? Yes, in the sense that many good-performing 11-year-olds subsequently do less well. No, in the sense that the odds of a high-achieving 11-year-old doing well at 16 are far higher than the odds of a random pupil doing so well.
For me, these figures raise questions about egalitarian politics.
Start from the fact that there are systematic differences in achievement according to class and ethnicity even at a young age - 11. And let's assume - subject to the caveat in point 3 - that these differences are correlated (imperfectly) with lifetime attainment. Let's also posit that 11-year-olds lack full moral agency and so cannot be held wholly responsible for their test scores.
It then follows that people do badly in life (partly) through no fault of their own.
Is this just?
If so, what can be done about it?
There are two answers I'm not happy with.
One is the equality of opportunity proposal to spend more on educating black or deprived children, to equalize their chances of attainment. The problem here is that it requires massive differences in spending to equalize opportunity, because the link between educational spending and achievement is weak.
The other is New Labour's proposal to end child poverty. One problem with this is that the link between attainment and income is monotonic; kids from the richest quartile do better than those in the second-richest quartile. Abolishing child poverty, therefore, would not equalize differences in attainment. Another problem is that it doesn't tackle ethnic differences in attainment.
So, what (if any) are the answers?

Comments

[dearieme]

The answers? Don't try to answer silly questions and instead try to find questions worth asking. Don't begin by using terminology such as "deprived", the likely purpose of which is to attempt to preclude intelligent thoght. Don't present to us the conclusion "..people do badly in life (partly) through no fault of their own" as if it is not a case of the bleedin' obvious. Don't ask "Is this just?" without expecting a loud raspberry in reply and an enquiry of whether the Pope shits in the woods. Don't betray your grammar school heritage by using the barbarism "good-performing". Not up to your usual standards, Master D: stay behind after school and write 500 lines.

[Mark Wadsworth]

There are no easy answers, but maybe you're asking the wrong question.

Maybe it's one of those things that we just have to accept, because the huge cost of trying to fix it (and nobody knows how) is just not worth it.

[Shuggy]

"So, what (if any) are the answers?"

Let's face it - no-one knows.

If I've understood some of that data correctly, sending cohorts of chemistry, physics, and biology teachers into deprived areas might be an idea? Have a more prescriptive curriculum where taking three sciences is compulsory and limiting options to take crappy subjects? Dunno how it works in Engerland but here schools in deprived areas tend to do the opposite and have all but abolished core subjects - which hasn't worked very well at all.

[Mark Wadsworth]

Dearime beat me to it by a whisker.

[Mark Wadsworth]

I suppose one simple answer to the racial differences would be to only allow Black Caribeans to marry Chinese (and vice versa) and only allow Indians to marry Pakistanis (and vice versa). Whites seem to be about average anyway. I doubt whether this would be a workable or popular measure though.

[Ami Ganguli]

Make sure that people aren't "channeled" to high or low paying jobs any earlier than absolutely necessary. That means ensuring that everybody has access to a quality university education at an affordable price, even if they just barely graduated high school.

Of course underachievers will likely continue to underachieve, but at least they will have had the opportunity to turn things around as adults.

[john b]

Mark W, that may be the best idea anyone has ever had, ever. You win awards.

John B

[Dr Dan H.]

What to do depends on what you want the outcome to be. From the figures, it looks like parental wishes are a lot more influence on young children than they are on older ones, where peer pressure is much more important.

So, if you want a well-rounded nation, try to select the high-achievers at age 10 or 12, and group them with peers that are similarly high achievers, so later on peer pressure keeps them going.

Act also to separate out the troublemakers, the stupid and the disruptive; you don't want these to act as role models for the bulk of the kids, and you don't need them in class since they're so disruptive to lessons.

Finally, don't be afraid to fail in a small percentage of cases. Some kids are unteachable and unmanageable; you'll never get good outcomes all of the time. Act instead to sequester the failures off out of the way of the bulk of the kids to let the majority get a good training.

[donpaskini]

There are several other options, surely? One is lots more post-16 education, so that people can pick up education and skills when they want to and are motivated to get the most out of it.

Another is to think about the implications of the link you suggest between lifetime attainment and educational achievement - narrowing the financial gap between high attainers and lower attainers so that high attainers have more incentives to pursue careers which aren't closely link to what will make them the most money.

I'd have thought that your argument is less of a challenge for egalitarians, and more difficult for the people who are interested in 'meritocracy'.

[TStockmann]

I've always been confused about the implication of phrases like "moral agency," which presupposes some quality of will that sounds - at best - metaphorical. For that matter the notion of whether endowments - genetic or environmental - are fair, since they're not "earned" - again begging the same question about moral agency. It also suggests that contestants in an Olympic footrace ought not be measured against each other directly, but rather rewarded to the extent their speed exceeds the speed one would expect them to exhibit, all things considered.

[Iain Coleman]

> It also suggests that contestants in an Olympic footrace ought not be measured against each other directly, but rather rewarded to the extent their speed exceeds the speed one would expect them to exhibit, all things considered.

This does happen in some sports: consider weight categories in boxing, for example. And even in athletics, competitors are generally segregated into men's and women's events in recognition of the physiological differences between the sexes.

[TStockmann]

A good point, but also an illustration of what I was describing - the arbitrariness and superficialty of the distinctions that are made - in this case two out of many thousands inborn or in opportunity to prepare (one could add the ban on performance enhancing drugs here) for the races.

[AJ]

This maybe the reason for racial differences. http://www.eastbayexpress.com/2003-05-21/news/rich-black-flunking/
What can you do about that?

[Sam]

Hypothesis:

Children who do well at school tend to come from homes where education is valued. People who value education tend to be better educated than people who don't, and tend to be better paid. There are cultural differences too (education tends to be highly valued in families of Chinese and Indian origin; children from those backgrounds tend to outperform their coaevals at school.)

Solution? For the white working class and black communities to adopt the attitude to education shown by the Indian immigrant couple who used to run the corner shop where I grew up. Two of their three children are doctors, and the third is an accountant.

The government can and should lead the children of the poor to the sweet, clear waters of education, but without support from the children's families, it's hard to persuade many of the children to drink.

[Jim]

"Abolishing child poverty, therefore, would not equalize differences in attainment."

But it would reduce the gap, which is worthwhile.

"Another problem is that it doesn't tackle ethnic differences in attainment."

To the extent that ethnic differences in attainment are due to ethnic differences in poverty and related issues, and it certainly looks like it has a role as it's the poorer ethnic groups that tend to have the worst attainment, surely reducing child poverty would tackle them. It would not eliminate them, but it would allow a clearer focus on the remaining issues, such as cultural factors. Ignoring the role of poverty is exactly what many of those who bang on about culture, absent dads etc as the cause of these differences want.

[Sam]

Is the problem actually poverty, or is poverty (partly) caused by a tendency not to value education?

[Chris P]

Without dropping into lachrymose sentimentality & waving my working class credentials perhaps ambition is missing. I left school at 15 & knew for a cast iron certainty that social mobility had to be sweated for through years of night school prior to university. Instead of attempting to make school 'relevant' & reacting to pupils points of view a confident school knows that they know better than their pupils what is good for them. They also know better than politicians as well. Taking up a point Dillow made earlier: If an independent Bank of England is a good thing perhaps an independent Dept of Education is also a good thing?

[Maynard Handley]

Easy. Just force everyone to have one and only one kid.

http://www.nytimes.com/2007/06/22/science/22sibling.html?ei=5090&en=727b74a8f5602af8&ex=1340164800&partner=rssuserland&emc=rss&pagewanted=all

This would also do a lot to help various environmental problems.

Next --- how about a *hard* question this time?